Research on prefecture-level city’s birth rate prediction based on AE-LSTM network

Abstract

In recent years, China's birth rate has continued to decline, leading to a series of population and economic problems. Although the country has gradually liberalized its Family Planning Policy and started implementing a Three-Child Policy with supporting fertility measures, it has not significantly increased the birth rate. To further implement the policy of encouraging childbirth at the prefecture-level city, this article proposes an AE-LSTM model for predicting the birth rate of the prefecture-level city, combining multidimensional time-series data that affect the birth rate of the population, which achieves accurate prediction of the birth rate of the prefecture-level city, providing a certain decision reference for prefecture-level city governments to formulate more accurate and reasonable childbirth encouragement policies. At the same time, experiment results show that the method proposed in this article has higher accuracy and better generalization performance compared with other mainstream methods.